Biochem. J. (1991) 279, 595-599 (Printed in Great Britain)

Purification and properties of kynurenine aminotransferase fromrat kidneyMadhumalti R. MAWAL, Arindam MUKHOPADHYAY and Devendra R. DESHMUKH*Department of Pediatrics, Children's Hospital of Michigan, Wayne State University, Detroit, MI 48201, U.S.A.

Previous reports indicated that a single protein exhibits kynurenine aminotransferase (KAT) and z-aminoadipateaminotransferase (AadAT) activities. However, recently we discovered that KAT and AadAT activities are associatedwith two different proteins. KAT from rat kidney supernatant fraction was purified to electrophoretic homogeneity by(NH4)2SO4 fractionation, DEAE-Sephacel and hydroxyapatite chromatography. This procedure separated KAT fromAadAT and improved the overall yield and the degree of purification over previously published methods. Some of theproperties of purified KAT, such as Mr, subunit structure and the inhibition by dicarboxylic acids, were identical withthose reported previously. However, the substrate specificity and pl of purified KAT were different from earlier reports.The same procedure can also be used to purify KAT from rat kidney mitochondria. These results support our earlierobservation that KAT and AadAT activities are associated with two proteins and suggest that cytosolic KAT may bestructurally similar to the mitochondrial enzyme.

INTRODUCTION

Kynurenine aminotransferase (KAT; EC 2.6.1.7), an enzymeof the tryptophan catabolic pathway, catalyses the irreversibletransamination reaction between L-kynurenine and a-oxo-glutarate to form kynurenic acid and L-glutamate. In mammals,KAT activity is present in various tissues such as liver, kidney,small intestine and brain and in different subcellular fractionssuch as mitochondria and cytosol [1-4]. The mitochondrial KATalso catalyses the transamination of 3-hydroxykynurenine intoxanthurenate [5]. oc-Aminoadipate aminotransferase (AadAT;EC 2.6.1.39), an enzyme ofthe lysine catabolic pathway, catalysesthe reversible transamination reaction between a-aminoadipateand a-oxoglutarate to produce a-oxoadipate plus L-glutamate[3,4].

Earlier methods of KAT purification were time-consuming,provided low yield and could not separate KAT from AadAT[1-3]. In all these studies, both activities co-purified with aconstant ratio of specific activities, showed similar chromato-graphic profiles and exhibited identical patterns of inactivationby heat and by dicarboxylic acids. Based on these results, it wasconcluded that a single protein catalyses both activities [1-3].However, recently we discovered that the two activities can beseparated by affinity chromatography [6]. The results describedin the present paper demonstrate that the two activities can beseparated by DEAE-Sephacel column chromatography, con-firming that AadAT and KAT are associated with two distinctproteins. An improved procedure for purifying KAT from ratkidney supernatant fraction is described, and the properties ofthe purified enzyme are reported.

MATERIALS AND METHODS

L-Kynurenine sulphate, kynurenic acid, a-aminoadipic acid,a-oxoglutaric acid, pyridoxal 5'-phosphate, standard proteinmolecular-mass markers, standard pl markers and otherchemicals were from Sigma Chemical Co., St. Louis, MO,U.S.A. Adult male albino rats (Sprague-Dawley) weighing

250-300 g were purchased from Charles River Laboratories,Wilmington, DE, U.S.A.

Enzyme assaysKAT activity was assayed as described by Knox [7]. The

reaction mixture contained 100 mM-imidazole/HCI, pH 6.5,3.3 mM-kynurenine sulphate, 5 mM-a-oxoglutarate and 0.2 mM-pyridoxal 5'-phosphate in a total volume of 0.6 ml. The assaymixture was incubated for 30 min at 37 °C, and the reaction wasterminated by adding 10 ml of 1% boric acid in 95% (v/v)ethanol. a-Oxoglutarate was added to control tubes after ter-mination of the reaction. Denatured protein was removed bycentrifugation at 5000 g for 10 min, and the A333 correspondingto kynurenic acid was measured against boric acid in ethanol.AadAT was assayed as described by Nakatani et al. [8]. Proteinconcentration was determined by the protein dye-binding assayof Bradford [9], with BSA as a standard.

Purification of KATRats were killed by decapitation and their kidneys were

immediately removed. All operations were carried out at 0-4 °C,unless stated otherwise. The kidneys were homogenized (10%,w/v) in 0.25 M-sucrose containing 0.1 mM-EDTA and 10 mm-Tris, pH 7.4. The homogenate was centrifuged at 600 g for10 min and the pellet was discarded. To separate mitochondria,the supernatant was centrifuged at 12000 g for 10 min. Thepurity of mitochondrial and supernatant fractions was checkedby assaying glutamate dehydrogenase (EC 1.4.1.3) and lactatedehydrogenase (EC 1.1.1.27) activities respectively [10,11].The procedure for purifying KAT from rat kidney supernatant

fraction is summarized in Table 1. Initial steps of purification,including acid precipitation and (NH4)2SO4 fractionation, werecarried out as described by Tobes & Mason [2]. The pelletobtained after (NH S)2504 fractionation was suspended in 8 mm-potassium phosphate buffer, pH 6.2, containing 10 mM-2-mercaptoethanol and 0.05 mM-pyridoxal 5'-phosphate (bufferA). The sample was dialysed for 20 h against the same buffer andloaded on the DEAE-Sephacel column (1 cm x 14 cm) pre-equilibrated with buffer A. The column was eluted with buffer A,

Vol. 279

595

Abbreviations used: KAT, kynurenine aminotransferase; AadAT, a-aminoadipate aminotransferase.* To whom correspondence should be addressed.

M. R. Mawal, A. Mukhopadhyay and D. R. Deshmukh

Table 1. Purification of KAT from rat kidney supernatant fraction

Details of the purification procedure are described in the text. One unit of activity is defined as the amount of enzyme that catalyses the formationof 1 ,umol of product/min at 37 'C.

Total activity Total protein Specific activity Recovery PurificationStep (units) (mg) (units/mg) (%) (fold)

SupernatantAcid precipitation(NH4)2SO4DEAE-SephacelHydroxyapatite

5.794.904.133.142.28

64033172.73.360.54

0.0090.0150.0570.944.22

10085715439

1.76.3

104469

and 1.0 ml fractions were collected (Fig. 1). The fractionscontaining KAT activity were pooled, dialysed for 20 h against8 mM-potassium phosphate buffer, pH 7.0, containing 10 mM-2-mercaptoethanol and 0.05 mM-pyridoxal 5'-phosphate (buffer B)and subjected to hydroxyapatite column chromatography. Thehydroxyapatite column (1 cm x 15 cm) was equilibrated withbuffer B and washed with 2 column vol. of the same buffer. Thecolumn was eluted with a linear gradient of 8-210 mM-potassiumphosphate buffer, pH 7.0, and 1.2 ml fractions were collectedand assayed for KAT activity. The active fractions were pooled,dialysed against buffer B, concentrated by Centricon- 10 (Amicon)filtration and used for further studies. A280 was used to identifyprotein peaks during DEAE-Sephacel and hydroxyapatitecolumn chromatography.

Polyacrylamide-gel electrophoresisTo check the purity of KAT, non-gradient polyacrylamide-gel

(7.5 %) electrophoresis was carried out at pH 8.3 [12]. The Mr ofthe subunits was estimated by gel electrophoresis in SDS [13].The enzyme was denatured by treatment with 1 % SDS solutioncontaining 1 % 2-mercaptoethanol at 100 °C for 5 min. BSA(Mr 66000), ovalbumin (Mr 45000), glyceraldehyde-3-phosphatedehydrogenase (Mr 36000), carbonic anhydrase (Mr 29000) andtrypsinogen (Mr 24000), used as Mr markers, were treated in thesame way.

Mr of the native enzymeA Sephadex G-200 column (1.5 cm x 46 cm) was equilibrated

with 50 mM-potassium phosphate, pH 7.0, containing 10 mM-2-mercaptoethanol [14]. The column was calibrated with thefollowing standard Mr markers: thyroglobulin (Mr 670000),fl-globulin (Mr 158000), ovalbumin (Mr 44000), myoglobin(Mr 17000) and vitamin B-12 (Mr 1350). Proteins were elutedwith the same buffer, and the elution was monitored by measuringthe A280. A plot of fraction number against log Mr was used todetermine the Mr of purified KAT.

Isoelectric focusingAadAT activity was purified as reported previously [6]. Iso-

electric focusing of purified KAT and AadAT was carried out on7.5 %-acrylamide gels containing 2% ampholyte and 10%sucrose, carrier ampholytes in the pH range 5-8 being used. Thegels were focused at 4800 V-h, fixed in 10% (w/v) trichloroaceticacid containing 5 % (w/v) sulphosalicylic acid and stained for10-12 h with 0.2% Coomassie Blue R-250. The gels weredestained with ethanol/acetic acid/water (4:1:5, by vol.).

Substrate specificityThe substrate specificity of purified KAT was determined by

using a number of a-amino acids as amino donors and witheither a-oxoglutarate or pyruvate as amino-group acceptor. The

E

21-

KAT

0 3 6 9 12 15Fraction no.

Fig. 1. Chromatographic separation of KAT and AadAT

A partially purified enzyme obtained after (NH4)2S04 fractionationwas dialysed and loaded on a DEAE-Sephacel column(1 cm x 14 cm), and 1.0 ml fractions were collected. Details ofchromatography are described in the text. Enzyme activities areexpressed as units/ml.

transamination reaction with aromatic amino acids was carriedout by the method of Takada & Noguchi [15].

Amino acid analysisAmino acid analysis was carried out at the protein-sequencing

facility of Wayne State University, Detroit, MI, U.S.A. Thepurified (unoxidized) sample was freeze-dried and subjected tovapour-phase hydrolysis at 150 °C for 90 min. The sample wasthen treated with phenyl isothiocyanate, and the derivative wasused for amino acid analysis by the Waters Pico-tag system [16].The loss of serine (17 %) and threonine (13 %) during hydrolysiswas corrected for accordingly.

Absorption spectraThe absorption spectrum for purified KAT (200 ,tg/ml) was

obtained at pH 5.0 and 8.2 in the presence of 0.05 mM-mercapto-ethanol [17].

RESULTS

Purification of KATThe mitochondrial fraction from rat kidney did not contain

any detectable lactate dehydrogenase activity, and the super-natant fraction was devoid of glutamate dehydrogenase activity.These results confirmed that the mitochondrial and supernatantfractions were not contaminated by other subcellular com-ponents. The purification procedure for KAT is summarized inTable 1. On DEAE-Sephacel column chromatography, KATactivity separated from AadAT activity (Fig. 1). During this

1991

596

Rat kidney kynurenine aminotransferase

Table 2. Substrate specificity of KAT purified from rat kidney1 0 3

x M,:.....

66 ::.::-:...... .......

In all assays 10 ,ug of purified KAT and 50 mm of either a-oxoglutarate or pyruvate was used. The final concentrations were:kynurenine, 3.3 mM; aromatic amino acids, 2.0 mM; DL-2-amino-pimelic acid, 80 mM; and other amino acids, 40 mm. Enzymeactivities were assayed as described in the Materials and methodssection and are expressed as the percentage of activity obtained withkynurenine and a-oxoglutarate as substrates. The data shown areaverages of duplicate analyses.

Amino-group donor a-Oxoglutarate Pyruvate

4:...e..-

2-4-

Fig. 2. SDS/PAGE of KAT

Purified enzyme (30 ,ug) was electrophoresed on 10 %-polyacrylamide gels in presence of 1% SDS, as described in theMaterials and methods section. The direction of migration was top(cathode) to bottom (anode). Lane 1, standard Mr markers: BSA(Mr 66000), ovalbumin (Mr 44000), glyceraldehyde-3-phosphatedehydrogenase (Mr 36000), carbonic anhydrase (Mr 29000) andtrypsinogen (Mr 24000). Lane 2, KAT.

Kynurenine5-HydroxykynurenineHistidineAminoadipateDL-2-Aminopimelic acidL-NorleucinePhenylalanine, tyrosine,tryptophan

100 460 815 110 01 51 4

<8 <4

Table 3. Effect of dicarboxylic acids on KAT activity purified from ratkidney

Purified KAT (10 ug) was preincubated with dicarboxylic acid (finalconcn. 6 mM; neutralized to pH 7.0) at 37 °C for 30 min in imidazolebuffer, pH 6.5. Enzyme activity was then determined as described inthe Materials and methods section. The data shown are averages ofduplicate analyses.

pI

Dicarboxylic acid added Inhibition (%o)7.2-

NoneDecanoic acidAdipic acidDiethylglutaric acidPimelic acidAzelaic acidDimethylglutaric acid3-Methylglutaric acid,glutaric acid, oxaloaceticacid

5.4

1 2 3

Fig. 3. Isoelectric focusing of KAT

Purified enzyme (30 ,tg) was electrophoresed on 7.5%-polyacrylamide gels as described in the Materials and methodssection. Lane 1, purified KAT. Lane 2, standard pl markers:myoglobin from horse heart (pl 7.2), myoglobin from horse heart(pl 6.8), carbonic anhydrase I from human erythrocytes (pl 6.6),carbonic anhydrase II from bovine erythrocytes (pl 5.9), carbonicanhydrase II from bovine erythrocytes (pl 5.4). Lane 3, AadATpurified from rat kidney supernatant fraction.

step, aspartate aminotransferase, which is usually associatedwith AadAT and KAT activities, was completely removed. TheKAT activity was purified 470-fold, with an overall yield of40%. The same procedure could also be used to purify KATfrom rat kidney mitochondria (results not shown).

Purity, M, and subunit structure

The purified KAT migrated as a single band on native PAGE,indicating that the preparation was homogeneous. On SephadexG-200 column chromatography, the purified enzyme was elutedbetween ,-globulin (Mr 158000) and ovalbumin (Mr 44000),indicating the Mr of 91000 (mean of two observations). The Mrof the subunit structure, estimated by PAGE in the presence of1% SDS, was found to be about 44000+ 1200 (n = 3; Fig. 2).These results indicate that KAT is composed of two apparentlyidentical subunits.

Isoelectric focusing and kinetic parametersThe pl of KAT was found to be 5.9 + 0.07 (Fig. 3). The Km

values for kynurenine and a-oxoglutarate were 4.0 mm and0.02 mm respectively.

Substrate specificityThe ability ofKAT to catalyse transamination between various

amino acids and keto acids is shown in Table 2. Maximumactivity was obtained with kynurenine as an amino-group donorand a-oxoglutarate as the amino-group acceptor. A significantactivity was also observed with 5-hydroxykynurenine as an

Vol. 279

45-

.636U..-....

6.8 -

66-

0414030232112<3

597

M. R. Mawal, A. Mukhopadhyay and D. R. Deshmukh

Table 4. Amino acid composition of KAT purified from rat kidney

Purified KAT was hydrolysed in vacuo at 150 °C for 90 min andthe amino acids were analysed as described in the text.

ContentAmino acid (residues/molecule)

Glutamic acidAspartic acidSerineGlycineHistidineArginineThreonineAlanineProlineTyrosineValineMethionineIsoleucineLeucinePhenylalanineLysine

A

124.082.086.075.011.830.539.488.034.629.359.010.030.846.838.039.8

Wavelength (nm)Fig. 4. Absorption spectra of KAT

Curve I represents the absorption spectrum ofKAT in 0.1 M-acetate,pH 5.2, and curve II is that in 0.05 M-Tris, pH 8.2.

amino-group donor and a-oxoglutarate as amino-group ac-ceptor. No measurable activity was observed when either a-aminoadipate or aromatic amino acids were used as amino-group donor and a-oxoglutarate or pyruvate as the amino groupacceptor. The purified KAT exhibited some (< 15 %) activitywith histidine as an amino-group donor and either pyruvate ora-oxoglutarate as amino-group acceptor (Table 2).

Effect of dicarboxylic acids and metabolitesSeveral dicarboxylic acids were tested as inhibitors of KAT

from kidney supernatant (Table 3). Adipic acid and decanoicacid caused the maximum inhibition. Dimethylglutaric acidcaused about 30% inhibition. However, oxaloacetic acid and 3-methylglutaric acid had no significant effect on the KAT activity.The effect of various metabolites of the lysine and tryptophanpathway on the KAT activity was studied. Aminoadipate andother metabolites did not cause any change in KAT activity(results not shown).

Amino acid analysisThe amino acid composition of purified KAT is shown in

Table 4. The purified enzyme had a high content of aspartic acid,glutamic acid, serine and alanine as compared with other amino

acids, whereas histidine and methionine were present in lowamounts. The high content of glutamic acid and aspartic acidpresumably reflects a substantial contribution from the hydrolysisof glutamine and asparagine respectively, since the pl valueindicates only a moderate excess of acidic over basic amino acidside chains.

Absorption spectraAt pH 5.2 KAT exhibited two absorption maxima at 330 nm

and 400 nm, whereas at pH 8.2 only one absorption maximum at330 nm was observed (Fig. 4).

DISCUSSION

Earlier studies indicate that KAT and AadAT, purified fromrat liver mitochondria as well as from rat kidney supernatantfractions, have a pl of 6.6 [2,3]. However, we found that the plvalues ofKAT and AadAT purified from rat kidney supernatantfraction were 5.9 and 6.6 respectively (Fig. 3). In previousstudies, a broader range of ampholytes (pH range 3-10) wasused, whereas we used a narrow range of ampholytes (pH 5-8).The isoelectric-focusing data are consistent with our observationthat two activities are resolved on DEAE-Sephacel chromato-graphy at pH 6.2 (Fig. 1). KAT and AadAT activities from ratkidney mitochondria were also separated on DEAE-Sephacelcolumn chromatography at pH 6.2 (results not shown). Theseresults indicate that KAT and AadAT activities from mito-chondrial fractions represents two distinct proteins and that thepI of the mitochondrial KAT may be the same as that ofsupernatant enzyme.

In the previous studies, aspartate aminotransferase activitywas associated with both AadAT and KAT activities until thelast stage of purification, and even the purified KAT/AadATexhibited a small amount of aspartate aminotransferase activity[2]. In our studies, aspartate aminotransferase activity wascompletely removed during DEAE-Sephacel column chromato-graphy. Therefore additional steps of purification after hydroxy-apatite column chromatography used in other studies were notnecessary.The Mr and subunit structure of the KAT purified from rat

kidney supernatant fraction were identical with those of enzymespurified from other sources [1-4,18]. KAT from rat kidneysupernatant fraction also resembled that from other sources inthe pattern and the degree of inhibition by dicarboxylic acids(Table 3).Although there are several reports on the purification ofKAT

from different sources, the amino acid composition of the enzymehas not yet been determined. The amino acid composition of ratkidney KAT (Table 4) was similar to that ofAadAT from bovinekidney [18], except that the numbers of lysine and prolineresidues in rat kidney enzyme were less than those in bovinekidney AadAT. High levels of glutamic acid and aspartic acidobserved in our study may be due to the hydrolysis of glutamineand asparagine, because the Pico-tag method used in our studydoes not distinguish between aspartic acid and asparagine andbetween glutamic acid and glutamine.The absorption spectrum of KAT (Fig. 4) resembles that of

AadAT and of most other aminotransferases in that it absorbslight in the wavelength range of 300-500 nm. At low pH (5.2) oneabsorption maximum appeared at 415 nm, which can be ascribedto a phosphopyridoxal-aldimine enzyme complex, and thesecond maximum was observed at 335 nm. At high pH (8.2)only one absorption maximum, at 335 nm, was observed. Thesespectra resemble those of aspartate aminotransferase from pigheart [17] and are characteristic ofpyridoxal phosphate-requiringenzymes.

1991

598

Rat kidney kynurenine aminotransferase 599

The Km values for kynurenine and a-oxoglutarate for thepurified enzyme in our studies were comparable with thosereported [1-3]. However, the substrate specificity of KAT fromrat kidney supernatant (Table 2) was different from that reportedpreviously. For example, previous studies indicated that, inaddition to kynurenine, other amino acids such as DL-3-hydroxykynurenine, DL-a-aminoadipate, DL-aminopimelate, L-norleucine and L-tryptophan also serve as amino donors, and a-oxoglutarate serves as amino-group acceptor [2,18,19] for KATactivity. However, we did not observe any activity with amino-adipate, norleucine or DL-aminopimelate as amino group donorsand with a-oxoglutarate as amino-group acceptor (Table 2).Aminoadipate was reported to be a competitive inhibitor ofKAT activity partially purified from rat kidney supernatantfraction and also from rat liver mitochondria [1,3]. However, wedid not observe any inhibition of KAT activity on addition ofaminoadipate. This difference could be due to removal ofAadATactivity from our preparations.

Recently we separated KAT from AadAT activity from ratkidney supernatant fraction [6]. However, the method was time-consuming, provided a low yield and required antibodies againstAadAT. the present method is rapid, separates KAT fromAadAT, and provides improved yield and degree of purificationover previously published methods.

This work was supported by a grant from the Children's Hospital ofMichigan, Detroit, MI, U.S.A.

REFERENCES1. Tobes, M. C. & Mason, M. (1975) Biochem. Biophys. Res. Commun.

62, 390-3972. Tobes, M. C. & Mason, M. (1977) J. Biol. Chem. 252, 4591-45993. Takeuchi, F., Otsuka, H. & Shibata, Y. (1983) Biochim. Biophys.

Acta 743, 323-3304. Hartline, R. A. (1985) Methods Enzymol. 113, 664-6725. Nakatani, M., Morimoto, M., Noguchi, T. & Kido, R. (1974)

Biochem. J. 143, 303-3106. Mawal, M. R. & Deshmukh, D. R. (1991) J. Biol. Chem. 266,

2573-25757. Knox, W. E. (1953) Biochem. J. 53, 379-3858. Nakatani, Y., Fujioka, M. & Higashino, K. (1970) Biochim. Biophys.

Acta 198, 219-2289. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254

10. Schmidt, E. (1974) in Methods of Enzymic Analysis (Bergmeyer,H. U., ed.), 650-659, Academic Press, New York

11. Kornberg, A. (1955) Methods Enzymol. 1, 441-44312. Davis, B. J. (1964) Ann. N.Y. Acad. Sci. 121, 404-42713. Weber, K., Pringle, J. R. & Osborn, M. (1972) Methods Enzymol.

26c, 3-2714. Andrews, P. (1965) Biochem. J. 96, 595-60615. Takada, Y. & Noguchi, T. (1987) Methods Enzymol. 142, 273-27916. Tarr, G. E. (1986) in Microcharacterization of Polypeptides: A

Practical Manual (Shirley, J. E., ed.), pp. 155-194, Humana Press,Clifton, NJ

17. Martinez-Carrion, M. & Tiemeier, D. (1967) Biochemistry 6,1715-1722

18. Deshmukh, D. R. & Mungre, S. M. (1989) Biochem. J. 261, 761-76819. Asada, Y., Sawa, Y., Tanizawa, K. & Soda, K. (1986) J. Biochem.

(Tokyo) 99, 1101-1110

Received 27 February 1991/18 June 1991; accepted 25 June 1991

Vol. 279